The applicant has previously developed a 3-dimensional viewing technology. This technology uses in preferred embodiments two or more overlapping liquid crystal display (LCDs) screens positioned parallel to but spaced apart from each other. In addition to the more obvious benefits of 3-dimensional displays in terms of realistic portrayal of depth, the applicant's display technology provides additional potential benefits in terms of preattentive information processing.
The term preattentive processing has been coined to denote the act of the subconscious mind in analysing and processing visual information which has not become the focus of the viewer's conscious awareness.
When viewing a large number of visual elements, certain variations or properties in the visual characteristics of elements can lead to rapid detection by preattentive processing. This is significantly faster than requiring a user to individually scan each element, scrutinising for the presence of the said properties. Exactly what properties lend themselves to preattentive processing has in itself been the subject of substantial research. Colour, shape, three-dimensional visual clues, orientation, movement and depth have all been investigated to discern the germane visual features that trigger effective preattentive processing.
Researchers such as Triesman [1985] conducted experiments using target and boundary detection in an attempt to classify preattentive features. Preattentive target detection was tested by determining whether a target element was present or absent within a field of background distractor elements. Boundary detection involves attempting to detect the boundary formed by a group of target elements with a unique visual feature set within distractors. It maybe readily visualised for example that a red circle would be immediately discernible set amongst a number of blue circles. Equally, a circle would be readily detectable if set amongst a number of square shaped distractors. In order to test for preattentiveness, the number of distractors as seen is varied and if the search time required to identify the targets remains constant, irrespective of the number of distractors, the search is said to be preattentive. Similar search time limitations are used to classify boundary detection searches as preattentive.
A widespread threshold time used to classify preattentiveness is 200-250 msec as this only allows the user opportunity for a single ‘look’ at a scene. This timeframe is insufficient for a human to consciously decide to look at a different portion of the scene. Search tasks such as those stated above maybe accomplished in less than 200 msec, thus suggesting that the information in the display is being processed in parallel unattendedly or pre-attentively.
However, if the target is composed of a conjunction of unique features, i.e. a conjoin search, then research shows that these may not be detected preattentively. Using the above examples, if a target is comprised for example, of a red circle set within distractors including blue circles and red squares, it is not possible to detect the red circle preattentively as all the distractors include one of the two unique features of the target.
Whilst the above example is based on a relatively simple visual scene, Enns and Rensink [1990] identified that targets given the appearance of being three dimensional objects can also be detected preattentively. Thus, for example a target represented by a perspective view of a cube shaded to indicate illumination from above would be preattentively detectable amongst a plurality of distractor cubes shaded to imply illumination from a different direction. This illustrates an important principle in that the relatively complex, high-level concept of perceived three dimensionality may be processed preattentively by the sub-conscious mind.
In comparison, if the constituent elements of the above described cubes are reorientated to remove the apparent three dimensionality, subjects cannot preattentively detect targets which have been inverted for example. Additional experimentation by Brown et al [1992] confirm that it is the three dimensional orientation characteristic which is preattentively detected. Nakaymyama and Silverman [1986] showed that motion and depth were preattentive characteristics and that furthermore, stereoscopic depth could be used to overcome the effects of conjoin. This reinforced the work done by Enns Rensink in suggesting that high-level information is conceptually being processed by the low-level visual system of the user. To test the effects of depth, subjects were tasked with detecting targets of different binocular disparity relative to the distractors. Results showed a constant response time irrespective of the increase in distractor numbers.
These experiments were followed by conjoin tasks whereby blue distractors were placed on a front plane whilst red distractors were located on a rear plane and the target was either red on the front plane or blue on the rear plane for stereo colour (SC) conjoin tests, whilst stereo and motion (SM) trials utilised distractors on the front plane moving up or on the back plane moving down with a target on either the front plane moving down or on the back plane moving up.
Results showed the response time for SC and SM trials were constant and below the 250 msec threshold regardless of the number of distractors. The trials involved conjoin as the target did not possess a feature unique to all the distractors. However, it appeared the observers were able to search each plane preattentively in turn without interference from distractors in another plane.
This research was further reinforced by Melton and Scharff [1998] in a series of experiments in which a search task consisting of locating an intermediate-sized target amongst large and small distractors tested the scrial nature of the search whereby the target was embedded in the same plane as the distractors and the preattentive nature of the search whereby the target was placed in a separate depth plane to the distractors.
The relative influence of the total number of distractors present (regardless of their depth) verses the number of distractors present solely in the depth plane of the target was also investigated. The results showed a number of interesting features including the significant modification of the response time resulting from the target presence or absence. In the target absence trials, the reaction times of all the subjects displayed a direct correspondence to the number of distractors whilst the target present trials did not display any such dependency. Furthermore, it was found that the reaction times in instances where distractors were spread across multiple depths were faster than for distractors located in a single depth plane.
Consequently, the use of a plurality of depth/focal planes as a means of displaying information can enhance preattentive processing with enhanced reaction/assimilation times.
Although not restricted to the exclusive use of same, LCD screens are particularly suited for use with the applicant's display technology. As is well known in the art, LCD screens are typically configured with crossed polarisers on either side of the liquid crystals layer. This configuration would therefore block the passage of light through two or more successive LCD screens arranged with identical polarisation axes.
In one means of addressing this problem, the applicants place a bi-refringent film between adjacent LCD screens. This film transforms the polarisation of the light coming from the rear LCD screen from linear polarisation to elliptical polarisation, before it passes through the front LCD screen. This enables the viewer to see an image displayed on the rear LCD screen through the front screen as the effect of the orthogonal polarisers is overcome.
However, further problems occur with the introduction of this extra element. Bi-refringent films do not have a uniform thickness even though the variance is only in the range of micrometres. Therefore, the different wavelengths of incident light such as that comprising ‘normal’ white light results in polarisation ellipsoids of different axial ratio and/or tilt angle, after transmission through the aforementioned bi-refringent film. Therefore, varying amounts of light of different wavelengths will pass through the polarisers of the front LCD, and the user will see bands of colour.
A further problem of viewing one LCD screen through another LCD screen is that the viewer's perception of the electronic tracery pattern used to address each of the pixels on the LCD screen. The combination of viewing the tracery on the back screen overlaid with the tracery on the front screen causes moiré interference patterns which are very noticeable and unwelcome to the viewer.
In order to address the above-described problems, the applicants formed the bi-refringent film with optically diffusive properties, by etching a matt surface onto one side of the film. This remedies the visual colour anomalies, and the viewer's perception of the tracery on the rear screen. Thus, it can be seen that the introduction of a diffuse element to this technology plays a pivotal role in producing an optically usable multi-LCD screen display.
It is important that the diffuse bi-refringent element is located close to the rear screen to provide the viewer with the greatest divergence of light and thereby providing a wider viewing angle.
A necessary effect of the diffusive nature of the bi-refringent film is a slight blurring of the image on the rear LCD screen. The optimum level of diffusion is the minimum amount that renders the appearance of moiré interference patterns invisible or insignificant. Further diffusion or blurriness is not only unnecessary but detrimental to the perceived quality of the image on the rear screen and hence that of the combined multi-screen display system.
The blurring of the rear screen is compounded by the way that LCD manufacturers presently manufacture their screens. At present, almost all LCD manufacturers produce screens that have a matt surface on one or both sides. This matt surface is intended to reduce glare on the LCD screen by randomly scattering the light reflected off the front of the LCD screen, so that a mirror-type reflected image is not perceived by a viewer. This matt finish may also be on the rear surface of the LCD screen to help to diffuse the screen's normal backlight source.
Unfortunately, this also increases the diffusion of the light being emitted by the rear screen when this standard LCD screen, with matt finish applied to the front and/or rear surface, is used with the applicant's technology as described. This results in an unnecessarily blurred rear image thereby restricting the utility of the 3-dimensional display.
It is an object of the present invention to address the foregoing problems or at least to provide the public with a useful choice.
Further aspects and advantages of the present invention will become apparent from the ensuing description which is given by way of example only.